This repository contain the caricature images with corresponding landmark in paper: Alive Caricature from 2D to 3D.

The Caricature_w_landmark folder contains the caricature images with its landmarks. We named them by number. e.g. 1.jpg with its landmarks file named 1.txt.

Each text file was organized by x-coordinate and y-coordinate of 68 landmarks. The sequence of landmarks following ibug_facial_landmarks.jpg

We provide a python script to read the landmarks file and display them.

Usage: run read_landmarks.py in this folder. For example, you want to show the 7-th image's landmarks, you can use:

cd ../Caricature_w_landmark
python read_landmarks.py --index 7

Dependencies: PIL.

It should be something like this. Those red points are facial landmarks.

The my_result_Crop folder contains outputs given by our method.

Add landmarks index of 3D model: Add 3D model landmark file 3D_landmark used in our project. Those data was marked manually and can be edited if necessary.

If you have comments or questions, please contact Qianyi Wu ([email protected]) and Juyong Zhang ([email protected]).

Please note that this dataset could be used for research purpose only, and any commercial use of the data is prohibited.

Please cite the following paper if this dataset helps your research: Alive Caricature from 2D to 3D

@article{wu2018alive
      title={Alive Caricature from 2D to 3D},
      author={Wu, Qianyi and Zhang, Juyong and Lai, Yu-Kun and Zheng, Jianmin and Cai, Jianfei}
      journal={CVPR},
      year={2018}
}